Poor Man's Pultec EQP1-A Build Support Thread

[funkymonksf]

Anyone finish there's? It'd be great to see some photos!

 

[funkymonksf]

What type switch is needed for the eq in&out? I assumed double pole, double throw. Wanted to double check. Also is that the same type for power on-off switch?

 

[wave]

What type switch is needed for the eq in&out? I assumed double pole, double throw. Wanted to double check. Also is that the same type for power on-off switch?

DPDT for EQ in/out. You can use an SPST for the power switch

Dave

 

[funkymonksf]

On the psu I see that I need a 100nf film with 250 "ac".. Is this correct? If so is there a special type of cap that is necessary or will any film work assuming it'll handle 250v? Part #?

Thanks,
Brice

 

[ruffrecords]

On the psu I see that I need a 100nf film with 250 "ac".. Is this correct? If so is there a special type of cap that is necessary or will any film work assuming it'll handle 250v? Part #?

Thanks,
Brice

This is connected across the 240V ac HT winding hence the 250V ac rating. You need remember this means the peak voltage across this cap will be over 350V dc so you could use any film cap rated at 400V dc. However, there are filem caps designed especially for this job and rated at 250V ac or 275V ac.  A goo choice would be this one:

http://uk.farnell.com/vishay-roederstein/f17724102000/capacitor-100nf-275v/dp/1612243

Cheers

Ian

 

[wave]

Ian,
So I got my hands on some cool surplus 18 pos switches and I'm going to make stepped pots for everything. I saw the post on page one about this (in 12 steps). My question is: when using the resistor calculator what values in db should I use? EJ_whyte used -40 for his max attenuation (first step) on his 470K log. Is that a correct place to start? Could it be more?
Please be aware that this is the first time I have attempted something like this and I'm still a noob in general (you know that though).

One more thing, on the mid boards thread you mentioned the three position switch for the boost/cut selector. Do you know of a part number for a switch like that I could use for reference?

Thanks,
Dave

p.s. I'm still planning on finishing the hookup diagram I've just had to put it on the back burner for a little while 

 

[ruffrecords]

Ian,
So I got my hands on some cool surplus 18 pos switches and I'm going to make stepped pots for everything. I saw the post on page one about this (in 12 steps). My question is: when using the resistor calculator what values in db should I use? EJ_whyte used -40 for his max attenuation (first step) on his 470K log. Is that a correct place to start? Could it be more?
Please be aware that this is the first time I have attempted something like this and I'm still a noob in general (you know that though).

Even or 18 steps, -40 would be a good place to start. The calculator should then divide that into equal steps.

One more thing, on the mid boards thread you mentioned the three position switch for the boost/cut selector. Do you know of a part number for a switch like that I could use for reference?

Thanks,
Dave

Any of these will do:

http://uk.farnell.com/jsp/search/browse.jsp?N=2031+203716+110145882+110177139&Ntk=gensearch&Ntt=spdt+toggle&Ntx=mode+matchallpartial&No=0&getResults=true&appliedparametrics=true&locale=en_UK&divisionLocale=en_UK&catalogId=&skipManufacturer=false&skipParametricAttributeId=&prevNValues=2031+203716+110145882&mm=1001953||,1002918||,&filtersHidden=false&appliedHidden=false&autoApply=false&originalQueryURL=%2Fjsp%2Fsearch%2Fbrowse.jsp%3FN%3D2031%2B203716%26Ntk%3Dgensearch%26Ntt%3Dspdt%2Btoggle%26Ntx%3Dmode%2Bmatchallpartial%26No%3D0%26getResults%3Dtrue%26appliedparametrics%3Dtrue%26locale%3Den_UK%26divisionLocale%3Den_UK%26catalogId%3D%26skipManufacturer%3Dfalse%26skipParametricAttributeId%3D%26prevNValues%3D2031%2B203716

You just need a spdt On-Off-On toggle switch.

Cheers

Ian

 

[wave]

Thanks Ian,
I want equal steps for a linear pot but the log pots need a curved response, right?

This will work for the switches, right?

http://www.mouser.com/ProductDetail/NKK-Switches/M2113TCFW01-RO/?qs=gndqSz9ohutnKt0mR5dl7oTaLWHsOuSQ%2f%2fZ9J67D6Xk%3d


Dave

 

[ruffrecords]

Thanks Ian,
I want equal steps for a linear pot but the log pots need a curved response, right?

Spot on.

This will work for the switches, right?

http://www.mouser.com/ProductDetail/NKK-Switches/M2113TCFW01-RO/?qs=gndqSz9ohutnKt0mR5dl7oTaLWHsOuSQ%2f%2fZ9J67D6Xk%3d


Dave

They look fine to me. I assume the illumination is just on all the time.

Cheers

Ian

 

[wave]

Ian,
I'm looking at those switches as they are red in one position and green in the other. No illumination in the off position.

Is there any guidelines for what curve to use for the log response? There is a chart on the Goldpoint website for 24 position but I need to break that down to 16

http://www.goldpt.com/r_series.html

Dave

 

[ruffrecords]

Ian,
I'm looking at those switches as they are red in one position and green in the other. No illumination in the off position.

Is there any guidelines for what curve to use for the log response? There is a chart on the Goldpoint website for 24 position but I need to break that down to 16

http://www.goldpt.com/r_series.html

Dave

OK, the first thing to remember is that all these sites and calculators are for making volume controls. As often as not they make the attenuation steps equal until you reach the end when of course they turn right off.

That's not quite what you want in an equaliser. For a start, the range of boost and cut is less than 20dB. We are more interested in the law of the pot so we get equal dBs of boost of cut per step rather than its attenuation in dB. A regular log pot attenuates by 20dB (one tenth) at the mid point . This means 90% of its value is from the wiper to the top and 10% from the wiper to the bottom. You should arrange your log switches to do the same but this corresponds to half the boost cut or around 10dB.

Confused? You should be because I am.

The bottom line is there is I am not sure if there is an easy and obvious way to design a stepped equaliser switch using programs/techniques/formulae intended for volume controls. What I really need to do is run the simulation of the EQ and work out the resistor values for 1dB and 2dB steps of boost or cut or for 10 or 20 steps all of which is quite a bit of work.

Anyway, I have just spent a few minutes going over the original design thread where ej white first suggested using stepped pots and it seems we did agree that it was best to set the range to 40dB on the basis that equal steps must make the mid point at 20dB as we require.

What we nned is a calculator to work out the values for us and I just found one here:

http://www.quadesl.com/attenuator.html

Just put in the resistance in ohms (470000 for 470K) select the number of steps and set the final attenuation as 40dB. Use the fixed logarithmic slope. I just ran this for 470K with 18 steps and rounding to the nearest resistor values I get:

1. 110K
2. 86K
3. 68K
4. 47K
5. 39K
6. 27K
7. 22K
8. 15K
9. 12K
10. 10K
11. 7K5
12. 5K6
13. 4K3
14. 3K3
15. 2K4
16. 2K
17.1K5


Hope that helps.

Cheers

Ian

 

[wave]

Ian,
Thanks! I actually came across this calculator in my quest for info on this project but it didn't work in Firefox on my mac.
I had to use Chrome on my Parallels PC (sigh).

So for the record, I should be cool with using the calculator on that site to find the values for my log "pots" and then just dividing the total resistance by the number of steps for my linear "pots"

Since I'm doing the stepped switches I have the opportunity to use the pot values in your design as opposed to the US vales (i.e. 5K, 50K, 500K)
Does it really matter which pot values I use provided I compensate via changing the capacitor values (and resistor) using your capcalc spreadsheet?
I ask as I have already painstakingly sourced the needed components based on the US pot values.

Once again, thanks for all your help!
Dave

 

[ruffrecords]

Ian,
Thanks! I actually came across this calculator in my quest for info on this project but it didn't work in Firefox on my mac.
I had to use Chrome on my Parallels PC (sigh).

So for the record, I should be cool with using the calculator on that site to find the values for my log "pots" and then just dividing the total resistance by the number of steps for my linear "pots"

Yes.

Since I'm doing the stepped switches I have the opportunity to use the pot values in your design as opposed to the US vales (i.e. 5K, 50K, 500K)
Does it really matter which pot values I use provided I compensate via changing the capacitor values (and resistor) using your capcalc spreadsheet?
I ask as I have already painstakingly sourced the needed components based on the US pot values.

The exact pot values are not critical so long as they remain in the same ratio to each other and you change the capacitor values appropriately.

Cheers

Ian

 

[wave]

Ian,
So would the switch be wired up so that the max resistance  (i.e 470K) is at the CW or CCW position?
Conversely, does the answer repeat for all the pots (i.e. is the answer is CW then the 4.7K Lin switch exhibits the max resistance in the CW position as well)

Thanks for your help!
Dave

 

[ruffrecords]

Ian,
So would the switch be wired up so that the max resistance  (i.e 470K) is at the CW or CCW position?

Max resistance should be with the 470K (lo cut) pot at the fully CW position.

Conversely, does the answer repeat for all the pots (i.e. is the answer is CW then the 4.7K Lin switch exhibits the max resistance in the CW position as well)

Thanks for your help!
Dave

The direction of rotation of all for pots is shown on the attached schematic (which is also included in the first post of this thread).

I would advise you to download and print all the documents listed in the first post.

Cheers

Ian

 

[wave]

Ian,
Thanks for the info. I actually have already downloaded every document and pic throughout both threads.
I think maybe I'm over thinking this.
Maybe I need to start with understanding the relationship of the info I get from the calculator to the assembling of the switch.
(I apologize for the penny not dropping yet!)
So according to the numbers you posted above:
1. 110K
2. 86K
3. 68K
4. 47K
5. 39K
6. 27K
7. 22K
8. 15K
9. 12K
10. 10K
11. 7K5
12. 5K6
13. 4K3
14. 3K3
15. 2K4
16. 2K
17.1K5

It all adds up to roughly 470K, but would step #1 be the first step rotating CW from a starting point of full anti-CW? Or would that be backwards?

Confused as ever,
Dave

 

[ruffrecords]

It all adds up to roughly 470K, but would step #1 be the first step rotating CW from a starting point of full anti-CW? Or would that be backwards?

Confused as ever,
Dave

I am not surprised you are confused. Once again, unfortunately, this calculator is for a volume control so it is not immediately obvious how you use it for EQ.

Anyway, to cut a long story short, with it fully clockwise we want about 470K and when it has been turned from there to the half way point we want about one tenth of that and sure enough if we add up the values from 9 thru 15 we get just over 50K. So, as you rightly guess it is backwards. We want it backwards which means you wire the 110K between positions 18 and 17, the 86K between positions 17 and 16 and so on (where position 18 is the fully clockwise position).

Hope that helps!!!

Cheers

Ian

 

[rmaier]

This goes a long ways to explaining why I've been having such problems getting mine going! I've wired everything up bass-ackwards and didn't take an attenuation of 40dB into account, using Harpo's calculator instead. Obviously, my values were very different. Ooops....

Before I make up another eight switches, I wonder if someone could check this attachment and see if I understand this correctly. I'm using 12-pos. switches, and the calculator Ian has suggested. Will this work?

Ralph

[Edit: removed attachment]

 

[wave]

I'm probably not the person who should be chiming in on this but looking at your pdf, if viewing the switch from the rear shouldn't your pin 1 be on the left side?

As far as I'm understanding it, the chart you have seems correct but like I said....

Dave

 

[wave]

Did I mention that my 18 pos switches are concentric? I also found some seriously cool old Daka-Ware concentric knobs




Dave